We are presently conducting a double-blind, randomized clinical trial in which more than 200 children with a history of at least one febrile seizure (FS) are given oral diazepam (DZP) or placebo at times of fever. Drug compliance is monitored; antipyretic measures, including all medicines taken, are noted; all FS recurrences and any afebrile seizures are recorded. We seek to answer the question: "Can oral DZP, taken only with fevers, prevent recurrences of FS as effectively as medication taken daily, and at lower risk to the child?" We view as major strengths of the study: ability to recruit and enroll patients; remarkably comparable randomized groups; close follow-up of enrolled subjects; minimal attrition of patients; excellent rapport with families; ongoing education of families; steadily improving drug compliance; adequate numbers of FS recurrences; medication safety; cooperation of the medical community. We work closely with families to educate them about subtle signs of illness, and reinforce the need for frequent temperature checks and prompt administration of study medications. We have learned that medication compliance is not always possible, but we are confident that our current 50% compliance rate will continue. This is a request for two additional years of support, enabling recruitment of subjects until 1/31/91, and follow-up of all randomized subjects until 1/31/92, followed by a final six months for statistical analysis of data. Considerations of sample size, statistical power and confidence intervals on the magnitude of treatment effects indicate that this two-year extension will allow us to provide an unequivocal, statistically-precise answer to the important public health question addressed by this research.